[Intel-gfx] The whole round of i-g-t testing cost too long running time

Ville Syrjälä ville.syrjala at linux.intel.com
Wed Apr 16 18:08:48 CEST 2014


On Wed, Apr 16, 2014 at 05:50:20PM +0200, Daniel Vetter wrote:
> On 16/04/2014 17:42, Jesse Barnes wrote:
> > On Tue, 15 Apr 2014 19:17:59 +0200
> > Daniel Vetter <daniel.vetter at intel.com> wrote:
> >> Ok there are a few cases where we can indeed make tests faster, but it
> >> will be work for us. And that won't really speed up much since we're
> >> adding piles more testcases at a pretty quick rate. And many of these
> >> new testcases are CRC based, so inheritely take some time to run.
> > But each test should run very quickly in general; I think we have too
> > many tests that take much longer than they need to.  Adding some
> > hooks to the driver via debugfs may let us trigger specific cases
> > directly rather than trying to induce them through massive threading
> > and memory pressure for example.
> >
> > And can you elaborate on the CRC tests?  It doesn't seem like those
> > should take more than a few frames to verify we're getting what we
> > expect...
> 
> Well they don't take more than a few frames, but we have a _lot_ of 
> them, and there's a lot of cominations to test. It adds up quickly. Iirc 
> we have over 150 kms_flip testcases alone ...

Many kms_flips tests are more like "run this thing for a few (dozen)
seconds and see if there's something unexpected reported". They don't
really try to hit specific issues.

Some of the more targeted crc tests do run much quicker, but I'm
guessing there are still at least two factors that add up. The actual
modesets that the tests have to do cost time. And then there's the
num_crtcs * num_connectors factor that tends to lengthen the execution
time considerably. If we could run more tests for each pipe/connector
pair w/o doing intervening modesets could probably cut the time down
somewhat. But that sounds like a fairly massive undertaking as all
the tests would have to be in a library and get called from some
generic test runner code that does the modeset part.

-- 
Ville Syrjälä
Intel OTC



More information about the Intel-gfx mailing list